Type 2 diabetes (T2D) is a major epidemic associated with significant burdens on patients, families, and the public healthcare system. Its rise in prevalence is concomitant with an increase in diabetes-related comorbidities. Among these, diabetes has emerged as a reproducible risk factor for cognitive impairment and dementia. However, the mechanisms underlying the risk for dementia in the disproportionately burdened T2D populations are poorly understood. The primary goal of this study is to evaluate the hypothesis that metabolomic signatures of neurocognitive trajectory are present in diabetes and these signatures explain, in part, race disparities in cognitive decline between European Americans and African Americans with T2D. This hypothesis will be explored by re-examining the Diabetes Heart Study (DHS) cohort for neurocognitive trajectory using a well-established cognitive battery, literacy testing, and adjudicated physician diagnosis of dementia. Because neurodegenerative conditions, representative of cognitive decline, are progressive with pathology developing years prior to the observation of clinical symptoms and functional deficits, untargeted metabolomic analysis will be performed on baseline samples collected >10 years prior and correlated with cognitive trajectory. This approach offers the potential to identify relevant biomarkers before onset of overt disease. Finally, a comprehensive genetic analysis of the DHS participants to examine the genetic architecture of neurocognitive measures and metabolomic signatures of neurocognitive change will be performed. The composition of this study, inclusive of European American and African Americans participants, will provide generalizability of the findings. The timing of this study is critical to contrast changes in midlife to early-late adulthood to identify first stage pathophysiological changes facilitating the identification of relevant biomarkers with potential to improve the diagnosis, prognosis and treatment of cognitive impairment and dementia.